Introducing: Comana

I've become really passionate about open source and finding ways to improve the effectiveness and sustainability of the communities built around these crucial projects. One of the things that might make this process easier is to have more readily available data to analyze. To that end I've been working on a side project for a couple weeks now called Comana.

Comana is an open source platform to provide an API to the data publicly available on GH Archive and parse it into actionable information. Right now it's a very simple serverless microservice hosted on AWS (ironic, I know) written entirely in Go. Currently, the API is only providing a simple count of GitHub events on an hourly basis but I'm looking to expand that functionality further.

• First, the project will be focusing on providing simple groupings of these metrics in daily, monthly, and yearly increments with an API that provides filtering options on the returned data
• Second, various statistics based off of these raw counts will be exposed. These could include growth/decline measurements, project comparisons/rankings, and percentage/delta changes over time.
• Third, and most ambitiously, the app could provide forward-looking projections. This is still very much TBD but it could range from regression to some basic basic machine learning.

Ideally, this app could also be expanded to include other communities like GitLab and BitBucket, but for now, GitHub is the initial target.

If you're interested in contributing:

• Respond to this thread with any questions/comments about the project design/goals
• Checkout the initial list of issues on the repo page to see what needs to be done
• DM me on Twitter for anything else

Happy coding!